Solving Dynamic Programming Problems on a Computational Grid (original) (raw)

Parallel Dynamic Programming on Clusters of Workstations

IEEE Transactions on Parallel and Distributed Systems, 2005

The standard DP (dynamic programming) algorithms are limited by the substantial computational demands they put on contemporary serial computers. In this work, the theory behind the solution to serial monadic dynamic programming problems highlights the theory and application of parallel dynamic programming on a general-purpose architecture (cluster or network of workstations). A simple and well-known technique, message passing, is considered.

Robust Asynchronous Optimization for Volunteer Computing Grids

2009

Volunteer computing grids offer significant computing power at relatively low cost to researchers, while at the same time generating public interest in different scientific projects. However, in order to be used effectively, their heterogeneity, volatility and restrictive computing models must be overcome. As these computing grids are open, incorrect or malicious results must also be handled. This paper examines extending the BOINC volunteer computing framework to allow for asynchronous global optimization as applied to scientific computing problems. The asynchronous optimization method used is resilient to faults and the heterogeneous nature of volunteer computing grids, while allowing scalability to tens of thousands of hosts. A work verification strategy that does not require the validation of every result is presented. This is shown to be able to effectively reduce the need for verification done to less than 30% of the reported results, without degrading the performance of the asynchronous search methods. An asynchronous version of particle swarm optimization (APSO) is presented and com- pared to previously used asynchronous genetic search (AGS) using the MilkyWay@Home BOINC computing project. Both search methods are shown to scale to MilkyWay@Home's current user base, over 75,000 heterogeneous and volatile hosts, something not possible for traditional optimization methods. APSO is shown to provide faster convergence to optimal results while being less sensitive to its search parameters. The verification strategy presented is shown to be effective for both AGS and APSO.

Job Allocation Schemes in Computational Grids based on Cost Optimization

19th International Parallel & Distributed Processing Symposium, Denver, Colorado, pp. 180-187, April 2005, 2005

In this paper we propose two price-based job allocation schemes for computational grids. A grid system tries to solve problems submitted by various grid users by allocating the jobs to the computing resources governed by different resource owners. The prices charged by these owners are obtained based on a pricing model using a bargaining game theory framework. These prices are then used for job allocation. We present the grid system model and formulate the two schemes as a constraint minimization problem and as a non-cooperative game respectively. The objective of these schemes is to minimize the cost for the grid users. We present algorithms to compute the optimal load (job) fractions to allocate jobs to the computers. Finally, the two schemes are compared under simulations with various system loads and configurations and conclusions are drawn.